Instantaneous Spectral Analysis: Time-Frequency Mapping Via Wavelet Matching With Application To 3D Gpr Contaminated Site Characterization

In recent years it has been demonstrated that time-frequency analysis, or spectral decomposition, can differentiate small-scale features associated with hydrocarbon reservoirs in seismic reflection data. Similar reflectivity anomalies are sometimes induced in ground-penetrating radar (GPR) data by electric property variations caused by groundwater contaminants that are often below the conventional resolution of the signal. Isolating and mapping discreet components of the time-frequency spectrum using spectral decomposition can highlight details of a contaminant distribution. The windowed fourier transform was an early approach to spectral decomposition, however wavelet based approaches have superior time localization properties. Here, we give the wavelet matching spectral decomposition algorithm we developed at the Houston Advanced Research Center in the mid 1990s. In a 3D GPR dataset acquired at the former Wurtsmith AFB, MI, the time-frequency attributes image details of a hydrocarbon plume not resolved by conventional instantaneous attributes or GPR AVO attributes.